Digital signal regulation system

Abstract

A system is disclosed for regulating a digital signal which is modulated such that an average of their amplitude over a given interval varies from interval to interval, in a manner to maintain the average constant. The digital signal comprises a train of pulses formed by pulse tops and pulse bottoms both of which are equally spaced from a reference level and at least one of the pulse top and pulse bottom being greater than two pulse widths. Signals which are modulated in accordance with MFM, RZ and NRZ systems correspond to such a digital signal. In accordance with the invention, those pulse tops or the pulse bottoms of the digital signal which have a selected one of the pulse widths are maintained unchanged in amplitude from its original value while other pulse tops or bottoms having different pulse widths are compressed in inverse proportion to their respective pulse widths. The system of the invention produces correction pulses for algebraic addition to those portions having the different pulse widths. The correction pulses have pulse widths which are equal to those of said other pulse tops or bottoms and have an amplitude which is determined in accordance with the pulse width thereof.

What is claimed is: 
     
       1. A system for regulating an average of amplitudes of a digital signal comprising a train of pulses formed by pulse tops and pulse bottoms which are equally spaced by a given amplitude from a reference level, at least one of the pulse top and the pulse bottom being a pulse variety which is of more than two different pulse widths; the system comprising means for detecting at least the presence of pulses of the pulse variety having a selected one of the pulse widths, means for producing correction pulses having a pulse width which is equal to said selected pulse width in response to the output of the detecting means, and means for algebraically adding the correction pulses with said pulses of the digital signal having the selected pulse width. 
     
     
       2. A system according to claim 1 in which the digital signal comprises a signal which is modulated according to a modified frequency modulation (MFM) system and in which the pulse top and the pulse bottom are each of three different pulse widths of T 0 , 3/2T 0  and 2T 0  where T 0  represents the period of a clock pulse and wherein the pulse tops and pulse bottoms are equally spaced by a given amplitude from a reference level, and wherein said detecting means detects pulses of the pulse variety which have the pulse widths of 3/2T 0  and 2T 0 , and said producing means produce correction pulses having amplitudes which are equal to one-third and one-half the given amplitude, and said adding means add the correction pulse having the one-third amplitude to a pulse of the pulse variety which has a pulse width of 3/2T 0  and adds the correction pulse having one-half amplitude to a pulse of the pulse variety which has a pulse width of 1/2T 0 . 
     
     
       3. A system according to claim 1 in which the digital signal comprises a signal which is modulated according to a modified phase modulation (MPM) system and in which the pulse top has a constant pulse width of 1/2T 0  while the pulse bottom has three different pulse widths of 1/2T 0 , T 0  and 3/2T 0  where T 0  represents the period of a clock pulse and wherein the pulse tops and pulse bottoms are equally spaced by a given amplitude from a reference level, and wherein the detecting means detect those pulse bottoms which have pulse widths of T 0  and 3/2T 0 , the producing means produce correction pulses having amplitudes which are one-half and two-thirds the given amplitude, and the adding means add the correction pulse having the one-half amplitude to a pulse bottom which has a pulse width of T 0  and adds the correction pulse having the two-thirds amplitude to a pulse bottom which has a pulse width of 3/2T 0 . 
     
     
       4. A system according to claim 1 in which the digital signal comprises a signal which is modulated to a modified frequency modulation (MFM) system and in which the pulse top and the pulse bottom are each of three different pulse widths of T 0 , 3/2T 0  and 2T 0  where T 0  represents the period of a clock pulse and wherein the pulse tops and pulse bottoms are equally spaced by a given amplitude from a reference level, and wherein the detecting means includes first circuit means for detecting the pulse tops having pulse widths of 3/2T 0  and 2T 0 , and second circuit means for detecting the pulse bottoms having the pulse widths of 3/2T 0  and 2T 0 , the first circuit means comprising means for producing a first train of position pulses indicative of the positions where the individual pulse tops of the pulse train end, a first gate receiving the first train of position pulses and detecting those first position pulses which correspond to the ending position of the pulse tops having the pulse width of 3/2T 0 , a first delay circuit for delaying the output of the first gate by 1/2T 0 , and a second gate receiving the first train of position pulses and detecting those first position pulses which correspond to the ending position of the pulse tops having the pulse width of 2T 0 , the second circuit means comprising means for producing a second train of position pulses indicative of the positions where the individual pulse bottoms of the pulse train end, a third gate receiving the second train of position pulses and detecting those second position pulses which correspond to the ending position of the pulse bottoms having the pulse width of 3/2T 0 , a second delay circuit for delaying the output of the third gate by 1/2T 0 , and a fourth gate receiving the second train of position pulses and detecting those second position pulses which correspond to the ending position of the pulse width of 2T 0 , and wherein said means for producing the correction pulses comprises a first and a second monostable multivibrator for producing correction pulses having a pulse width of 3/2T 0  in response to the outputs of the first and the second delay circuits, and a third and a fourth monostable multivibrator for producing correction pulses having a pulse width of 2T 0  in response to the outputs of the second and the fourth gate, and wherein said adding means comprises a delay circuit which delays the original pulse train by 2T 0 , and a matrix circuit for algebraically adding the outputs of the monostable multivibrators, with predetermined polarities, to the outputs of the delay circuit. 
     
     
       5. A system according to claim 1 in which the digital signal comprises a signal which is modulated according to a modified frequency modulation (MFM) system and in which the pulse top and the pulse bottom are each of three different pulse widths of T 0 , 3/2T 0  and 2T 0  where T 0  represents the period of a clock pulse and wherein the pulse tops and pulse bottoms are equally spaced by a given amplitude from a reference level, and wherein the detecting means comprises means for producing a combined train of position pulses including first position pulses indicative of the ending position of the pulse tops of the pulse train and second position pulses indicative of the ending position of the pulse bottoms, a first primary gate receiving the combined train of position pulses and detecting those first position pulses which correspond to the ending position of the pulse tops having a pulse width of 3/2T 0  and those second position pulses which correspond to the ending position of the pulse bottoms having a pulse width of 3/2T 0 , a second primary gate receiving the combined train of position pulses and detecting those first position pulses which correspond to the ending position of the pulse tops having a pulse width of 2T.sub. 0 and those second position pulses which correspond to the ending position of the pulse bottoms having a pulse width of 2T 0 , a first delay circuit for delaying the output of the first primary gate by 1/2T 0 , a first auxiliary gate for detecting those first position pulses which correspond to the ending position of the pulse tops having a pulse width of about 3/2T 0  in response to the output of the first delay circuit, a second auxiliary gate for detecting those second position pulses which correspond to the ending position of the pulse bottoms having a pulse width of 3/2T 0  in response to the output of the first delay circuit, a third auxiliary gate for detecting those first position pulses which correspond to the ending position of the pulse tops having a pulse width of 2T 0  in response to the output of the second primary gate, and a fourth auxiliary gate for detecting those second position pulses which correspond to the ending position of the pulse bottoms having a pulse width of 2T 0  in response to the output of the second primary gate, and wherein said means for producing the correction pulses comprise a first and a second monostable multivibrator for producing correction pulses having a pulse width of 3/2T 0  in response to the outputs of the first and the second auxiliary gate, and a third and a fourth monostable multivibrator for producing correction pulses having a pulse width of 2T 0  in response to the outputs of the third and the fourth auxiliary gate, and wherein said adding means comprises a second delay circuit which delays the original pulse train by 2T 0 , and a matrix circuit for algebraically adding the outputs of the monostable multivibrators, with predetermined polarities, to the outputs of the second delay circuit. 
     
     
       6. A system according to claim 1 in which the digital signal comprises a signal which is modulated according to a modified phase modulation (MPM) system and in which the pulse top has a constant pulse width of 1/2T 0  while the pulse bottom has three different pulse widths of 1/2T 0 , T 0  and 3/2T 0  where T 0  represents the period of a clock pulse and wherein the pulse tops and pulse bottoms are equally spaced by a given amplitude from a reference level, and wherein said detecting means comprises means for producing a train of position pulses indicative of the ending position of the pulse bottoms of the original pulse train, a first gate receiving the train of position pulses and detecting those position pulses which correspond to the ending position of the pulse bottoms having a pulse width of T 0 , a first delay circuit for delaying the output of the first gate by 1/2T 0 , and a second gate receiving the train of position pulses and detecting those position pulses which correspond to the ending position of the pulse bottoms having a pulse width of 3/2T 0 , wherein said means for producing the correction pulses comprises a first and a second monostable multivibrator for producing correction pulses having pulse widths of T 0  and 3/2T 0  in response to the output of the first circuit and the output of the second gate, and wherein said adding means comprises a second delay circuit which delays the original pulse train by 3/2T 0 , and a matrix circuit for algebraically adding the outputs of the monostable multivibrators to the output of the second delay circuit. 
     
     
       7. A system according to claim 1, wherein said means for producing correction pulses is effective for producing correction pulses having amplitudes such that when the correction pulses are added to the pulses of the digital signal the average amplitude of the sum of the digital signal and the correction pulses remains substantially constant.